Analysis of Accuracy in Identification of Bone Fracture using Canny Edge and Prewitt Edge Detection Approach

Abstract

In this study, we compare the New Modified Canny edge detection method to the Prewitt edge detection method to determine whether the method is more effective at identifying bone fractures. We will accomplish this by contrasting the two approaches. Methods and Materials: This study takes a ten-person sample and compares it to another ten-person sample using an innovative modified Canny edge detector (CED) and a Prewitt edge detector (PED). With the use of the g power software, we were able to compare our sample sizes using the following settings: alpha = 0.05, enrollment ratio = 0.1, 95% confidence interval = 80%, and power = 80%. The results of the study demonstrated that a customised version of the Canny edge detection method had an accuracy of 95% and a specificity of 86%. This result outperformed the Prewitt edge detection method in terms of accuracy and specificity. With an initial test statistical power of 80% in SPSS analysis and an accuracy of p = 0.006 (p 0.05) and specificity of p = 0.025 (p 0.05), it was determined that the data obtained left no room for error. The significance level was too low (p-value 0.05) to rule out this conclusion. Compared to the traditional Prewitt edge detection approach, the novel modified Canny edge detection method is significantly more accurate when diagnosing bone fractures.